US 7256899 B1 Abstract Methods for acquiring an approximation of the surface geometry of a 3-dimensional object include projecting pattern of structured light on the object, moving the pattern with respect to the object, acquiring images of the intersection of the light on the object over time, determining local coordinates of points on the intersection with respect to the pattern, tracking the position of the pattern, transforming the local coordinates to object coordinates, and accumulating the points as a model of the surface of the object. The methods include a step for wirelessly and possibly compactly transmitting geometrical data which characterizes an intersection for a given position of the scanner with respect to the object. Systems for embodying the method include a self-powered wireless, non-contact optical scanner, the location and orientation of which may be tracked with respect to an object coordinate system.
Claims(16) 1. A method for acquiring an approximation of a surface geometry of a 3-dimensional object comprising:
establishing an object coordinate system in known relationship to the object;
projecting a pattern of structured light of known geometry onto the object;
forming an image of an intersection of the pattern of structured light with the object;
processing the image to generate a set of data characterizing the intersection relative to a position of the pattern of structured light;
wirelessly transmitting some portion of the image and intersection data to a receiver;
receiving the transmitted portion of the image and intersection data;
tracking the position of the pattern of structured light;
associating each intersection datum with the position of the projected pattern of light at the time the image corresponding to the datum was formed;
transforming each intersection datum into coordinates of the object coordinate system; and
accumulating the transformed coordinates to form an approximation of the surface of the object.
2. The method of
3. The method of
4. The method of
5. The method of
6. The method of
7. The method of
8. The method of
9. The method of
10. The method of
11. The method of
12. The method of
13. The method of
14. The method of
15. The method of
16. A system for acquiring an approximation of a surface geometry of a 3-dimensional object comprising:
means for establishing an object coordinate system in known relationship to the object;
means for projecting a pattern of structured light of known geometry onto the object;
means for forming an image of an intersection of the pattern of structured light with the object;
processing means for generating a set of data characterizing the intersection relative to a position of the pattern of light;
transmitting means for transmitting some portion of the image or intersection data to a receiver; receiving means for receiving the transmitted processed intersection data;
tracking means for tracking the position of the projected pattern of structured light;
means for associating each intersection datum with the position of the projected pattern of light at the time the image corresponding to the datum was formed;
transforming means for transforming each intersection datum into coordinates of the object coordinate system; and
accumulating means for accumulating the transformed coordinates to form a model approximating the surface geometry of the object.
Description None The present invention relates to methods and systems for wireless, non-contact mensuration of the surface shape of a three-dimensional object or features thereof. Presently, computer graphics, 3D modeling of real-world objects in software, and 3D mechanical design are in widespread use. Accordingly, it is desirable to capture an approximate computer software model of an existing physical object. In many cases, it may suffice for the model to be a “point cloud” of surface points sampled from the physical object. In a more sophisticated model, the points are vertices of abutting planar polygonal patches which approximate the surface of the object. In an even more sophisticated model, the patches are curved, each defined by a bivariate polynomial or rational mathematical function, as in the NURBS surfaces commonly used in computer graphics. Special cases of 3D data entry also rely on parametric entry, where the geometric shape (circle, sphere, cube, etc.) of the object is known and the parametric features are specified in order to be retained during scanning. Numerous approaches exist which automate or partially automate the process of sampling representative points from the surface of an object. One approach generates a single point at a time, as a contact probe tip is moved across the surface of the object. Traditional coordinate measuring machines (CMMs) as well as handheld magnetic, mechanical, and optical probe tracking systems have historically used this contact probe approach. Computer-aided design (CAD) software can be used to accumulate the measured points and to build therefrom a 3-dimensional model of the surface of the object. An example of a handheld, optically tracked probe and its tracking system are the FlashPoint and 3D Creator products sold by Boulder Innovation Group, Inc. (Boulder, Colo.). Another approach might include various laser or optical beam non-contact probes, which operate similar to the contact probe approach. However, the “probe tip” is a narrow light beam together with an optical sensor, which accurately measures the length of the ray-like beam to where it intersects the object at a point. That distance together with knowledge of the exact position of the light beam allows computation of the 3-dimensional XYZ coordinates of the point where the beam intersects the surface of the object. By gathering sufficiently dense points from the object, software can create a suitable model of the surface of the object. To date, such non-contact probes are tethered at least by an electronic cable, if not by additional mechanical linkage. Rather than a single ray-like laser beam, more advanced non-contact scanners project a planar “fan” of light to illuminate many points on the object, where the intersection of the light and the object forms an illuminated stripe or contour line on the object. These scanners are sometimes called “laser stripe triangulation” scanners. One or more video cameras acquire a 2D dimensional image of the contour line from a position offset from the plane of light. In effect, the image of the contour line on each camera simultaneously captures the locations of many surface points all on one plane through the object. This speeds the process of gathering many sample points, while the plane of light (and usually also the receiving camera) is laterally moved so to “paint” some or all of the exterior surface of the object with the plane of light. By knowing the instantaneous position of the camera and the instantaneous position of the plane of light within a object-relative coordinate system when the image was acquired, a computer and software can use triangulation methods to compute the coordinates of illuminated surface points. As the plane is moved to intersect eventually with some or all of the surface of the object, the coordinates of more points are accumulated. A number of commercially available systems employ the technique of projecting a manually moveable plane of light and imaging the illuminated intersection of the light on the object. Examples include, the U.S. Pat. No. 4,737,032 (Cyberware, Inc., Monterey, Calif.), the FastSCAN (Polhemus Inc., Colchester, Vt.), and K-Scan (Metris N. V., Leuven, Belgium). Other systems project more than one plane of light onto the object, or project even more complex patterns of structured light. Examples are Metris's X-Scan product and U.S. Pat. No. 7,009,717 by Coppenolle et al, which is incorporated herein by reference. Other systems project a pattern of light that is modulated by sequentially increasing and decreasing the light line widths, sometimes using a Gray Code Sequence. One variation of this moves the line pattern over the object, referred to as “phase shift encoding”. Examples of these systems are GFM Messtechnik, ABW-3D, both of Germany, Opton (Japan), stereoSCAN by Breackmann and GOM of Germany. To date, such non-contact scanners are tethered at least by an electronic cable, if not by further mechanical linkage. For example, see U.S. Pat. No. 6,611,617 of Crampton. The present invention is directed toward a method and a system for wireless three dimensional shape sensing and mensuration. The method provides steps which generate a virtual 3-dimensional model approximating the surface geometry of the object. The system generates a virtual 3-dimensional model approximating the surface geometry of an object. The method projects a pattern of structured light on the surface of the object, acquires images of the intersection of the light with the object's surface, and processes each image to generate an intermediate set of surface data characterizing the intersection relative to the position of the pattern of light at the moment the image was acquired. The process tracks the position of the pattern of light with respect to the object over time. The surface data is wirelessly transmitted to a receiver which is connected to a computer which correlates each surface point datum with the temporally corresponding position of the pattern of light, transforms the surface point data into coordinates of points with respect to an object coordinate system. The transformed points are the basis of a virtual 3-dimensional model of the object. The system uses a 3-dimensional (3D) object coordinate system in a fixed or otherwise known relationship to the object, and the system uses a non-contact scanner. The non-contact scanner employs a source of a pattern of structured light, an electro-optical imager, an image processor, a wireless data transmitter, and a scanner position indicator. The non-contact scanner projects the structured light pattern onto the object. The light reflected from the illuminated intersection of the light pattern with the object at a particular moment is imaged by the electro-optical imager. The image processor generates surface point data which characterize the locations of points on the intersection relative to the scanner position indicator. The wireless transmitter transmits the image and/or surface point data wirelessly to a data receiver. The system also uses a scanner tracking subsystem which essentially continuously determines the 3D position (location and orientation) of the scanner position indicator, over time and with respect to the object. A computing module is in wired or wireless communication with the data receiver and to the scanner tracking subsystem. The computing module correlates each received surface point datum temporally with the position of the scanner position indicator at the time the intersection containing the surface point was imaged, transforms the received surface point data according to the correlated scanner positions, and generates a set of 3D surface point coordinates which are relative to the object coordinate system. The set of surface points are used to model approximately the object's surface. The method and the system assume that the position of the pattern of light and the position of the object are not necessarily fixed with respect to each other. Further, the method and the system assume that over time the pattern of light and the object move relative to each other to illuminate eventually a sufficient number of points on the surface of the object. The relative motion may be effected manually or automatically. The accompanying drawings, which are incorporated herein and constitute part of this specification, illustrate presently preferred embodiments of the invention, and, together with the general description given above and the detailed description given below, serve to explain features of the invention. Various embodiments of the present invention will be described in detail with reference to the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts. Note that herein the position of some (rigid) physical body means both its 3-dimensional location and its 3-dimensional orientation. As is common practice in computer science, the location of the body may be represented as the spatial XYZ coordinates of a specified reference point or marker on the body with respect to a defined Cartesian coordinate system. Further, the orientation of a body may be represented by yaw, pitch, and roll angles with respect to a defined reference orientation, by a quaternion, or by a conventional 3-by-3 rotation matrix. As is also common practice, the position may alternatively be represented by a single 4-by-4 homogeneous coordinate transformation matrix. As used herein, the terms “about” or “approximately” for any numerical values or ranges indicate suitable dimensional tolerances that allow the part or collection of components to function for their intended purposes as described herein. With reference to Wireless, non-contact scanner In the embodiment where the pattern is simply a single, static narrow ray-like beam, the beam is, for example, generated by a laser diode and a collating lens. The single, narrow ray-like beam intersects the object In alternative embodiments, the pattern of structured light is generated by projecting coherent light through a hologram or a diffraction grating to generate a planar “fan” of light, a Moiré fringe pattern, or a more complex set of multiple rays, planes, or curved surfaces in 3D space. Alternatively, the structured light can be generated by an optical projector such as a slide projector or digital video projector. In the case of a digital video projector, the geometry of the pattern of structured light may be dynamically varied so that different patterns are projected over time. In each of the embodiments, the pattern of structured light may be automatically moved across portions of the object without having to move the scanner itself. For the simple planar pattern of light Examples of systems projecting structured light include systems sold by GOM GmbH (Braunschweig, Germany) and GFMesstechnik GmbH (Berlin, Germany). In these systems the scanner is not hand-held, is placed at a fixed position, or is moved mechanically. At least one optical imaging sensor Unless specified otherwise, the following descriptions of embodiments will assume that the pattern of structured light may be composed of one or more rays, planes, or curved sheets of light. Then the image of an intersection In an embodiment, the image processing may discard some or all of the video pixels which are not part of the image of the illuminated intersection An embodiment may further transform the 2D subpixel coordinates of such a point on the image into 3D coordinates relative to a local Cartesian coordinate system defined with respect to the scanner In the case where the structured light pattern More than one lens and sensor Embodiments projecting any particular pattern of structured light The data generated by scanner The geometric data, in one or more of the above forms, is transmitted wirelessly via data transmitter U.S. Pat. No. 6,608,688, which is incorporated herein by reference, presents a tracking system which tracks wireless instruments such as probes. That patent describes one or more moveable trackable instruments each with position indicators, which are tracked in a coordinate space, and to transmit and receive various data wirelessly. In particular, the patent discloses techniques for synchronizing the tracking of multiple wireless instruments and for multiplexing the wireless communication with those multiple instruments. The geometric data sent by data transmitter The data transmitter A 2-dimensional frame of image pixels can be transmitted as a stream of many bytes (for monochrome) or many byte triplets (for color). The protocol may use some proprietary or standardized raw or compressed format, such as JPEG. If an embodiment transmits the pixel or subpixel coordinates of points on the image of the intersection At lease one trackable position indicator A source of energy (not shown in The scanner Scanner tracking subsystem In an embodiment using tiny or point-like position indicators The embodiment of For simplicity and clarity, the specification assumes that the pattern of structured light The computer The computer An embodiment may display the accumulated points or patches on a graphics screen In an embodiment, the object coordinate system Although Data transmitter Data transmitter Transmitter Handle Scanner Data record format Data record format Data record format Data record format Data record format Using the example formats of An embodiment method for acquiring an approximation of the surface geometry of a 3-dimensional object, without needing to contact the object physically, comprises steps listed in Step Step Step Step Step It may be most convenient to define the scanner coordinate system mentioned in step Step In an embodiment the times of capturing images in a scanner and the times at which the scanner position is determined may be asynchronous or at least may not closely match. In such an embodiment, the position of the scanner for a specific time (such as the time an image was captured) may be interpolated from two or more known positions of the scanner which are at times very close to the time of the image capture. For example, suppose the scanner position is known at two times t Step As the pattern of structured light is moved across the surface of the object, step Step Step Steps Step Step Step Step Using the geometrical information about the 3D relationship of the image to the pattern of light, step Once the coordinates of the intersection for any given position of the scanner relative to the object have been transformed into object (global) coordinates, the method of In an embodiment which allows the object being scanned to be moved so that the object (global) coordinate system moves with the object, the scanner may not need to be moved at all. The object itself may be moved and rotated so that the (potentially stationary) pattern of structured light eventually intersects with the entire surface of the object (or at least all parts of interest on the surface). See step While the present invention has been disclosed with reference to certain preferred embodiments, numerous modifications, alterations, and changes to the described embodiments are possible without departing from the sphere and scope of the present invention, as defined in the appended claims. Accordingly, it is intended that the present invention not be limited to the described embodiments, but that it have the full scope defined by the language of the following claims, and equivalents thereof. Patent Citations
Non-Patent Citations
Referenced by
Classifications
Legal Events
Rotate |